The purpose is to advance our knowledge of gastric cancer etiology with a multidisciplinary team approach to 2 unique populations. One in Narino, Colombia has one of the highest known risks. The others are the blacks of south Louisiana whose high risk contrasts with the rest of the U.S. and is so far unexplained and unexplored. Cohorts of individuals will be followed for 3 to 5 years. Gastroscopy and gastric biopsies will be used to classify precursor lesions of the gastric mucosa: atrophic gastritis, intestinal metaplasia, dysplasia. These will be combined with cell kinetics studies to evaluate the dynamics of the precancerous process and to correlate with environmental parameters and genetic markers. A search for carcinogens in gastric juice and nitrosated local foods will be done by a combinaion of mutagenesis and chemical techniques. Experiments to test the carcinogenicity of nitrosated local foods and the inhibition of carcinogenesis by antioxidants and other compounds will be conducted.